Towards accurate abdominal tumor segmentation: A 2D model with Position-Aware and Key Slice Feature Sharing

• Published in: Computers in biology and medicine 2024-09, Vol.179, p.108743, Article 108743
• Main Authors: He, Jiezhou, Luo, Zhiming, Lian, Sheng, Su, Songzhi, Li, Shaozi
• Format: Article
• Language: English
• Subjects: Abdomen; Abdominal Neoplasms - diagnostic imaging; Abdominal tumor segmentation; Algorithms; Datasets; Deep learning; Humans; Imaging, Three-Dimensional - methods; Information retrieval; Key slices; Position-Aware; Segmentation; Tomography, X-Ray Computed - methods; Transformer; Tumors; Two dimensional models
• ISSN: 0010-4825; 1879-0534; 1879-0534
• DOI: 10.1016/j.compbiomed.2024.108743

Abdominal tumor segmentation is a crucial yet challenging step during the screening and diagnosis of tumors. While 3D segmentation models provide powerful performance, they demand substantial computational resources. Additionally, in 3D data, tumors often represent a small portion, leading to imbalanced data and potentially overlooking crucial information. Conversely, 2D segmentation models have a lightweight structure, but disregard the inter-slice correlation, risking the loss of tumor in edge slices. To address these challenges, this paper proposes a novel Position-Aware and Key Slice Feature Sharing 2D tumor segmentation model (PAKS-Net). Leveraging the Swin-Transformer, we effectively model the global features within each slice, facilitating essential information extraction. Furthermore, we introduce a Position-Aware module to capture the spatial relationship between tumors and their corresponding organs, mitigating noise and interference from surrounding organ tissues. To enhance the edge slice segmentation accuracy, we employ key slices to assist in the segmentation of other slices to prioritize tumor regions. Through extensive experiments on three abdominal tumor segmentation CT datasets and a lung tumor segmentation CT dataset, PAKS-Net demonstrates superior performance, reaching 0.893, 0.769, 0.598 and 0.738 tumor DSC on the KiTS19, LiTS17, pancreas and LOTUS datasets, surpassing 3D segmentation models, while remaining computationally efficient with fewer parameters. •PAKS-Net: a 2D model for abdominal tumor segmentation in CT images, superior to 3D models.•Proposes KSFS, a module enhancing edge slice segmentation in 2D models by using adjacent context.•Proposes a PA module using positional correlation in CT images to improve tumor segmentation accuracy.